Abstract
Objective
Systemic lupus erythematosus (SLE) is a chronic, multisystem autoimmune disease characterized by periods of remission and recurrent flares, which have been associated with stress. Despite the significance of stress in this disease, the Perceived Stress Scale-10 has yet to be psychometrically evaluated in patients with SLE.
Methods
Exploratory factor analysis was used to examine the structural validity of the Perceived Stress Scale-10 among patients with SLE (N = 138) receiving medical care at Cedars Sinai Medical Center. Cronbach’s coefficient alpha was used to examine internal consistency reliability, and Pearson product-moment correlations were used to examine convergent validity with measures of anxiety, depression, helplessness, and disease activity.
Results
Exploratory factor analysis provided support for a two-factor structure (comparative fit index = .95; standardized root mean residual = .04; root mean square error of approximation = .08). Internal consistency reliability was good for both factors (α = .84 and .86). Convergent validity was evidenced via significant correlations with measures of anxiety, depression, and helplessness. There were no significant correlations with the measure of disease activity.
Conclusion
The Perceived Stress Scale-10 can be used to examine perceived stress among patients with SLE.
Systemic lupus erythematosus (SLE) is a chronic, multisystem autoimmune disease with a variable presentation and course. 1 Because multiple organs can be affected, patients with SLE present and continue with diverse symptoms including fatigue, weight loss, joint inflammation, skin rash, and a lower defense against infection. The timing of disease exacerbations, or flares, can appear random and unpredictable. The majority of cases of SLE occur in women, often starting in childbearing age. SLE is highly characterized by periods of remission and recurrent flares, which have been associated with stress. 2 Among patients with SLE, stress has been adversely associated with disease activity, quality of life, and morbidity. 2 Despite the significance of stress in this disease, the Perceived Stress Scale (PSS), a widely-used measure of perceived stress, has yet to be psychometrically evaluated in patients with SLE.
The PSS measures the degree to which situations in one’s life are appraised as stressful. 3 Three versions of the measure have been developed. The original measure was developed among two samples of college students (N = 446) and one community sample (N = 64), and consisted of 14 items, seven which were positively worded (e.g. “In the last month, how often have you felt that things were going your way?”) and seven items which were negatively worded (e.g. “In the last month, how often have you been upset because of something that happened unexpectedly?”). Response options range from 0 (never) to 4 (very often). Total scores are calculated by first reverse scoring the positively worded items, and then summing all items. Higher scores indicate greater perceived stress. The measure was later reduced to 10 items, removing four items with low factor loadings among a large national sample of adults living in the United States (N = 2387). 4 The measure developers recommended the 10-item version over the longer measure because it showed better structural validity and internal consistency reliability in the large national sample. 4 A four-item measure was also developed for situations requiring a very brief measure of perceived stress, and was originally used in telephone interviews for smoking cessation studies. 4 The present study will focus on the 10-item measure, the Perceived Stress Scale-10 (PSS-10), as it is the version of the measure recommended for use by the scale authors. 4
The PSS-10 has been translated into several languages, and the measure has been psychometrically evaluated in clinical (e.g. cardiac patients) and non-clinical (e.g. university students, policewomen) populations around the world (e.g. China, Mexico 5 ). A review examining the psychometric properties of the PSS-10 suggested that the two-factor structure is a better fit than a one-factor structure, although both factor structures have been supported. 5 In addition, internal consistency reliability has been good across studies (α values > .70). 5 Convergent validity has been demonstrated via significant relationships in the expected directions with measures of depression and anxiety. 5 Psychometric information on the measure is still needed, however, among patients with SLE. A measure should be demonstrated to be valid and reliable in a population prior to its use.
Thus, the present study evaluated the structural validity, internal consistency reliability, and convergent validity of the PSS-10 in a sample of patients with SLE. Based on previous studies, a two-factor structure was expected. 5 Internal consistency, as measured by Cronbach’s coefficient alpha, was expected to be good. Evidence of convergent validity was also hypothesized via significant, moderate to strong, positive correlations with the Hospital Anxiety and Depression Scale (HADS), 6 a measure of anxiety and depression, the Arthritis Helplessness Index (AHI),7,8 a measure of helplessness, and the SLE Disease Activity Index (SLEDAI), 9 a measure of disease activity. A review on the PSS noted that the measure has been moderately or strongly correlated with measures of anxiety and depression, including the HADS.5,6 The relationship between the PSS and the AHI and SLEDAI, however, have not been previously examined. A moderate to strong, positive correlation with the Arthritis Helpless Index was expected because helplessness is a symptom of depression. A moderate to strong, positive correlation with the SLEDAI was hypothesized because stress, as well as psychosocial stressors, has been associated with greater disease activity and worsened clinical symptomatology in SLE.10,11
Methods
Participants and procedures
This was a cross-sectional study of a non-probability sample of patients aged 18 years and above diagnosed with SLE (American College of Rheumatology 1982 criteria) from Southern California. Patients were recruited from the Cedars Sinai Medical Center, a private non-profit academic medical center and tertiary hospital serving a diverse catchment area. Patients with a previous diagnosis of psychiatric illness prior to SLE onset and those diagnosed with SLE for the first time during the clinic visit were excluded. Written informed consent to participate in the study was obtained. The study was approved by the Institutional Review Board at the Cedars Sinai Medical Center.
Measures
Patients completed study questionnaires while they received infusion or while waiting to be seen by a physician if they were not receiving infusions. Demographic and psychosocial measures were collected via patient self-report. SLE disease activity was collected by physicians.
PSS-10. 4
As described above, the PSS-10 is a self-report measure of perceived stress. The PSS-10 has six negatively worded items (e.g. been upset or angered) and four positively worded items (e.g. felt confident or in control of things). Total scores on the measure can range from 0 to 40, with higher scores indicating greater perceived stress.
HADS. 6
The HADS is a 14-item self-report screening instrument used to assess for risk of anxiety and depression. Two subscale scores can be calculated by summing individual items, one each for anxiety and depression. Subscale scores can range from 0 to 21, with higher scores indicating greater anxiety or depression. Internal consistency reliability in the present study was good for the anxiety (α = .86) and depression (α = .80) subscales.
AHI. 7
The five-item helplessness subscale of the AHI was used to measure helplessness, as recommended by Stein, Wallston, and Nicassio. 8 Response options range from 1 (strongly disagree) to 4 (strongly agree). A total score is calculated by summing items, and can range from 5 to 20. Higher scores indicate greater helplessness related to symptoms of arthritis. Internal consistency reliability in the present study was acceptable (α = .69)
SLEDAI. 9
The SLEDAI is the most widely-used measure of disease activity among patients with SLE. The 24-item measure includes an assessment of weighted clinical items and laboratory tests results, with more severe symptoms noted with higher scores. A total score is calculated, and can range from 0 to 105. Higher scores indicate greater disease activity.
Data analysis
Sociodemographic and clinical characteristics of the sample were calculated in SPSS version 23. The structural validity of the PSS-10 was evaluated in Mplus version 7.2 using exploratory factor analysis. As recommended by Bentler, 12 overall model fit was determined by consulting three descriptive fit indices: 1) the root mean square error of approximation (RMSEA 13 ) the standardized root mean residual (SRMR 14 ), and 3) the robust comparative fit index (CFI 15 ). For RMSEA and SRMR indices, values less than .08 were considered acceptable fit and values less than .05 were considered good fit. For CFI, values greater than .90 were considered acceptable fit and values greater than .95 were considered good fit. If values for two or more of the descriptive fit indices indicated at least acceptable model fit, the model was determined to fit well. The chi square (χ2) test of model fit was also reported for completeness, but not used as a primary indicator of model fit because it is sensitive to sample size. 15 Internal consistency reliability was examined using Cronbach’s coefficient alpha. Convergent validity was evaluated using Pearson product-moment correlation coefficients to examine the relationship of the PSS-10 to the HADS, AHI, and SLEDAI.
Results
Descriptive statistics
Sample characteristics (N = 136)
M (SD).
n (%).
HADS: Hospital Anxiety and Depression Scale; SLEDAI: SLE Disease Activity Index; AHI: Arthritis Helplessness Index
Structural validity and internal consistency reliability
Factor loadings of the Perceived Stress Scale-10
p < .05
Convergent validity
The Perceived Self-efficacy factor was significantly, positively, moderately associated with the HADS Anxiety (r = .29, p < .01) and Depression (r = .25, p < .01) subscales and the AHI Helplessness subscale (r = .25, p < .01). The Perceived Helplessness factor was significantly, positively, strongly associated with the HADS Anxiety (r = .71, p < .01) and Depression (r = .68, p < .01) subscales and the AHI Helplessness subscale (r = .50, p < .01). The Perceived Self-efficacy and Perceived Helplessness factors were not significantly associated with the SLEDAI (Self-efficacy: r = .06, p = .52; Perceived Helplessness: r = .07, p = .43).
Discussion
The present study’s findings suggest that the PSS-10 is a valid and reliable measure that can be used among patients with SLE. Exploratory factor analysis provided support for a two-factor structure, consistent with previous studies in other populations. 5 Based on findings from the present study, future research that uses the PSS-10 among patients with SLE should examine the two subscales of the measure, as opposed to the total score. In addition, internal consistency reliability was good for both subscales, providing further support for the two-factor structure. Additional psychometric validation of the PSS-10 is needed among patients with SLE to confirm the present study’s findings.
Participants’ scores on the Perceived Self-efficacy subscale indicated moderate to high levels of perceived self-efficacy. Participants’ scores on the Perceived Helplessness subscale indicated moderate to low levels of perceived helplessness. Mean subscale scores in the present study were similar to a previous study of predominantly female undergraduate students enrolled in public universities in the southeast United States. 16 A review of psychosocial research in SLE noted the limited research available comparing psychological distress across patients with SLE and healthy controls. 17 Furthermore, among studies that have examined this relationship, findings have been mixed. Studies are needed that compare levels of psychological distress in patients with SLE and individuals without disease. The PSS-10 can be used to facilitate this research, and potentially provide insight to psychological factors amenable to intervention to improve outcomes in patients with SLE.
Convergent validity was evidenced via significant correlations in the expected directions and magnitudes with measures of anxiety, depression, and helplessness. There were no significant correlations, however, with the measure of disease activity. A review examining the impact of stress on disease in patients with SLE found no causal relationship between stress and disease activity. 2 Rather, stress was identified as a moderator of disease activity. The non-significant relationship between the PSS-10 subscales and disease activity in the present study is consistent with these findings, and suggests that other factors play a more important role in disease activity among patients with SLE. Lazarus and Folkman highlight the importance of coping in response to stressors, and state that how an individual appraises and copes with an event determines the degree of stress caused by the event. 18 According to the goodness of fit hypothesis, coping will be most effective if the coping strategy used (e.g. problem focused, emotion focused) is appropriate for the stressor. Thus, coping styles utilized by patients, as opposed to the stressor itself, may play a more important role in disease activity in SLE. In addition, individual personality differences, for example in self-efficacy, have also been suggested to influence health outcomes and may impact disease activity in SLE. 10 In a previous study of 200 patients with SLE, lower self-efficacy for disease management was associated with greater disease activity, even when controlling for level of organ damage. 10
There were limitations to the present study. Patients participating in this study were a convenience sample, received treatment at a private hospital from doctors with expertise in SLE, and the majority were taking medication for SLE, limiting the generalizability of study findings. The present analysis did not consider how medication management, SLEDAI scores, or medical comorbidities may impact the psychometric properties of the PSS-10 because of the limited sample size. In addition, because this was a cross-sectional study, test–retest reliability and sensitivity to change could not be examined. Recognizing these limitations, this study provides support for the use of the PSS-10 to measure perceived stress among patients with SLE. Given the present limitations, future studies should attempt to replicate the psychometric validation of the PSS-10 in larger samples of patients with SLE to confirm the present study’s findings.
Footnotes
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: MHW is a consultant to UCB, Ionis Pharmaceuticals, and Ampel Biosolutions. He has contracted research with Human Genome Sciences, UCB, Eli Lilly and Company, Genentech, Department of Defense/Immunomedics, and EMD Serono. All other authors declare no conflicting interests.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.